JP2008241255A - Method of detecting alignment mark position and laser machining device using method thereof - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve accuracy of detecting an alignment mark position. <P>SOLUTION: A method of detecting the alignment mark position detects at least one alignment mark on an object to be positioned from an image obtained by picking up the object. The method includes a mark acquisition step of acquiring the alignment mark set up beforehand from the image, a reference point detection step of detecting four points as reference points from the alignment mark acquired in the mark acquisition step, a line segment acquisition step of acquiring a plurality of line segments from the reference points detected in the reference point detection step, and a position detection step of detecting the position of the alignment mark from the intersections of the line segments acquired in the line segment acquisition step. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an alignment mark position detection method and a laser processing apparatus using the method, and more particularly to an alignment mark position detection method and a laser processing apparatus using the method for improving the alignment mark position detection accuracy. .

  For example, in the field of laser processing or the like, in order to align the processing target held on the stage, the stage, a mask for forming an image in a desired shape when the processing target is irradiated with laser light, etc., at an appropriate position In addition, at least one alignment mark consisting of dots, crosses, stars, etc. is provided on the object to be processed, stage, mask, etc., and the area of the object including the alignment mark is photographed from a certain distance by a camera or the like. Then, the position of the alignment mark is detected from the photographed image, and the object is adjusted to an appropriate position.

  Here, as a method for detecting an alignment mark from an image obtained by photographing or the like, a template matching method for detecting the position by matching with a template of a preset alignment mark is known (for example, Patent Documents 1 and 2).

As another method, a projection method is known. Here, FIG. 1 is a diagram illustrating an example of a conventional projection method. The projection method shown in FIG. 1 integrates the luminance values of the image in the X-axis and Y-axis directions in the image with respect to the cross alignment mark 11 present in the image 10 obtained by photographing or the like, and the values are continuous. Then, the intermediate point of each of the X axis and Y axis of the section that is equal to or greater than the threshold is detected as the alignment position ((X, Y) coordinates).
Japanese Patent Laid-Open No. 06-151274 JP 2003-156111 A

  Incidentally, the methods disclosed in Patent Documents 1 and 2 perform matching by calculating the similarity between a template stored in advance and an alignment mark included in the current image. For this reason, for example, when the size of the object is large, the alignment mark near the end of the shooting area of the camera is deformed due to the angle between the object and the camera on the captured two-dimensional image. Displayed. Therefore, there is a problem that the alignment mark cannot be correctly detected when matching with the template.

  Further, in the template matching method, since the alignment mark is detected based on the similarity to the template, the position of the intersection cannot be accurately detected. Therefore, when acquiring the intersection, it is necessary to specify the position of the intersection on the template.

  In addition, the projection method described above has the following problems. FIG. 2 is an example for explaining a case where the projection method cannot be applied. As shown in FIG. 2, when the entire area of the alignment mark 11 does not fit in the image 10, as described above, the coordinates of the respective intermediate points in the section where the luminance value is continuously equal to or greater than the threshold value are the intersection points. Since the coordinates are not obtained, the position of the alignment mark cannot be detected accurately.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide an alignment mark position detection method for improving the alignment mark position detection accuracy and a laser processing apparatus using the method. .

  In order to achieve the above-described object, the present invention provides an alignment mark position detection method for detecting at least one alignment mark present on an object from an image obtained by photographing the object to be aligned. A mark acquisition step for acquiring a set alignment mark, a base point detection step for detecting at least four base points from the alignment mark acquired by the mark acquisition step, and a plurality of base points detected by the base point detection step A line segment acquisition step for acquiring the line segment, and a position detection step for detecting the position of the alignment mark from the intersection of the line segments obtained by the line segment acquisition step. As a result, the base point is detected from the alignment mark existing on the target object such as a mask or the like to be processed, the stage or the mask, and the detected base point is used to improve the position detection accuracy of the alignment mark. Can be improved.

  Further, the base point detection step is a point existing in a region constituting the alignment mark, the base points are separated from each other by a predetermined distance, and further, three or more points are not arranged on the same straight line. It is preferable to detect as Thus, by obtaining the base point, the position of the alignment mark can be detected even when all of the alignment marks are not included in the image.

  Furthermore, it is preferable that the base point detection step detects the base point from the contact portion when the alignment mark is a cross mark and there is a contact portion between the edge of the imaging region of the image and the alignment mark. Thereby, the base point can be easily detected. Further, even when not all of the alignment marks are included in the image, the position of the alignment marks can be detected.

  Further, in the base point detection step, when at least four base points are not detected, a part of the straight line constituting the alignment mark is searched by a preset search region, and the base point is detected from the search result. preferable. Thereby, the base point can be detected with high accuracy.

  Further, in the base point detection step, when three or more base points are detected on the same straight line, it is preferable to leave the points at both ends as base points and delete the other points. Thereby, since the distance between base points can be lengthened, the straight line which passes along two base points can be acquired correctly. Therefore, the position detection accuracy of the alignment mark can be improved using this straight line.

  Further, in the mark acquisition step, a search area set for searching for the alignment mark in the image is classified into any one of a mark area, a non-mark area, and a high content ratio area. It is preferable to be characterized by this. Thereby, since different searches can be performed in each region, alignment marks can be acquired quickly and accurately.

  According to the present invention, in the laser processing apparatus that performs processing by irradiating a laser beam to a positioned workpiece by the alignment mark position detection method, the optical system unit is moved in the optical axis direction with respect to the laser beam. Then, an optical system unit driving unit that irradiates the laser light imaged on the processing object, and scans the laser light from the optical system unit by moving a mirror to irradiate a predetermined position of the processing object. Image processing for detecting the position of the alignment mark with respect to the image picked up by the mirror driving means, the processing object and / or the surface of the stage holding the processing object, and the image picked up by the imaging means On the basis of the position detection result of the alignment mark obtained by the apparatus and the image processing apparatus. And having a control means for adjusting the position of that laser beam. Thereby, since the exact position of a target object can be grasped | ascertained, highly accurate laser processing is realizable.

  According to the present invention, it is possible to improve the alignment mark position detection accuracy. Thereby, since the exact position of a target object can be grasped | ascertained, highly accurate laser processing is realizable.

  DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a preferred embodiment of an alignment mark position detection method and a laser processing apparatus using the method according to the present invention having the above-described features will be described below in detail with reference to the drawings. In the following description, a cross shape is used as an example of the alignment mark 11. However, the present invention is not limited to this, and other shapes (marks) can be similarly handled.

<Detection of alignment mark position>
The position detection of the alignment mark according to the present embodiment is performed on, for example, a processing object made of resin or the like, a stage for holding the processing object, a mask for forming an image in a desired shape when the object is irradiated with laser light, or the like. The marked cross mark is photographed by an imaging means such as a camera, and the base point (coordinates) of the cross mark is detected from a captured image including the cross mark by image processing or the like, and the position (intersection point) of the cross mark is detected from the base point.

Here, FIG. 3 is a diagram for explaining a base point of the alignment mark in the present invention. In the present embodiment, the alignment mark 11 included in the image 10 defines the base point 12 on condition of the following three items (1) to (3).
(1) The point exists on a straight line constituting the cross mark.
(2) The base points are sufficiently separated from each other (when an image is quantized, if the base points are close to each other, a straight line or an inclination passing through the two base points cannot be detected).
(3) Three or more points are not arranged on the same straight line.

  That is, in the detection of the cross point of the cross mark in the present embodiment, for example, as shown in FIG. 3, first, at least four base points 12-1 to 12-4 are detected, and a linear expression 13 constituted by them is calculated. In FIG. 3, one straight line 13-1 is calculated from the base points 12-1 and 12-3, and one straight line 13-2 is calculated from the base points 12-2 and 12-4.

  Next, the position of the alignment mark 11 is detected by obtaining the intersection of the straight lines 13-1 and 13-2.

<Procedure for detecting cross point of cross mark>
Here, a cross mark detection procedure for the cross mark in the present embodiment will be described with reference to a flowchart. FIG. 4 is a flowchart illustrating an example of a cross mark detection procedure.

  In FIG. 4, first, an image for detecting a cross mark as an alignment mark is acquired (S01), and an area corresponding to the cross mark is acquired from the acquired image based on pixel information such as a luminance value (S02). .

  Next, it is determined whether or not the cross mark acquired in the process of S02 is in contact with the edge of the image area (S03). When the cross mark is in contact with the edge of the image area (YES in S03), a contact point is obtained from the contacted area, and the coordinates of the contact point are stored as a base point (S04).

  If the cross mark is not in contact with the edge of the image area (NO in S03), it is next determined whether or not at least four base points have been detected (S05). When at least four base points have not been detected (NO in S05), a part of a straight line (for example, mark end) constituting the cross mark is searched (S06), and the base point (coordinates) is detected from the search result. (S07).

  Further, a linear equation composed of a plurality of base points is calculated (S08). Further, it is determined whether or not there are three or more base points on the same straight line from the straight line (line segment) obtained by the straight line equation calculated in S08 (S09).

  Here, when there are three or more base points on the same straight line (YES in S09), the base points at both ends are left, the other base points are deleted (S10), the process returns to S05, and the processes after S05 described above are performed. Do. Further, in the process of S09, when there are not three or more base points on the same straight line (NO in S09), the process returns to S05, and the processes after S05 described above are performed. When the process of S06 is repeated, the straight lines searched in S06 are searched for straight lines excluding the straight lines searched up to the previous time.

  When at least four base points are detected in the process of S05 described above (YES in S05), a linear equation passing through the two base points is calculated (S11), and a straight line (line segment) obtained by the calculated linear equation is calculated. )), A combination in which the coordinates of the intersection of the two straight lines is within the image area is selected (S12), and the position of the cross mark for obtaining the intersection is determined (S13).

  With the processing procedure described above, the position of an alignment mark such as a cross mark can be detected with high accuracy based on the base point even when a part of the cross mark is off the screen.

  Next, a specific method for detecting the base point described above will be specifically described.

<Cross mark area detection method>
In detecting the base point, first, a cross mark area included in the target image is detected. FIG. 5 is a diagram illustrating an example of a cross mark region detection method. Usually, the alignment mark has a certain width. In the present embodiment, as shown in FIG. 5, for each axis (X, Y) of the image area, from a search section [a, (a + D)] that is narrower than the width of the band-shaped object that forms the straight line of the cross mark. Define a search area. Further, the luminance values are integrated by the projection method while moving the above-described search region at predetermined intervals with respect to the respective axes, and the values are acquired (a, a + 1, a + 2,...). In addition, the total number of pixels that exist in each search region and have a luminance value that exceeds a predetermined threshold is obtained, and the content rate α of the pixels that exceed the threshold in the search region is calculated.

Also, a range of pixels (La, Lb,...) That continuously exceed the threshold is obtained from the obtained projection data for each axis, and the maximum value _Lan and the second largest value L _bn among them are acquired. Is detected.

Further, from the change amount and the content of L _an, and L _bn obtained alpha, images within the search interval "(1) mark area, (2) non-mark area, (3) high content region" three Are classified into any one of them.

Here, (1) the mark area is an image area that is considered to include a part of the cross mark. When a result with a small amount of change of both L _an and L _bn or L _an is obtained in the entire search section, it is classified into this region, and it is considered that a cross mark is included in a portion where the amount of change is small. .

  Also, (2) the non-mark area is an image area that is regarded as not including a cross mark in the search section. A pixel value that is classified when the content rate is lower than the set value and exceeds the threshold value detected during that time is considered to be noise.

Furthermore, (3) the high content region is a region where the content rate α is high and cannot be ignored as noise, but since the amount of change in L _an and L _bn is large, it cannot be determined even if a cross mark is shown. Here, FIG. 6 is a figure for demonstrating a high content rate area | region. The high content area includes, in addition to an area where noise is large, a part of a cross mark (for example, a tip) where a base point cannot be detected as shown in FIG. 6A or a cross as shown in FIG. It is classified when a search area is set near the intersection of marks.

<Base point search and detection method>
Next, the base point search and detection method described above will be specifically described with reference to the drawings. FIG. 7 is a diagram for explaining the search and detection method of the base point. As shown in FIG. 7, first, it is determined whether or not the cross mark is in contact with the edge of the image area shot by applying the cross mark area detection method described above.

  Here, when the search area at the time of the search is classified into the mark area described above, as shown in FIG. 7A, on the central axis (x + D / 2 in FIG. 7) of the search section [x, (x + D)]. The center coordinates of the mark area (here, a section having a luminance value exceeding a predetermined threshold) is set as a base point.

  If the search area including the edge of the image area described above is classified as a high content area, the search area is shifted toward the center of the image as shown in FIG. Apply detection techniques. As a result, it is possible to prevent erroneous detection of noise having a uniform size as shown in FIG.

  This is repeated, and when the search area changes from the high content area to the mark area by a preset number of times, the base point is detected for the area using the method shown in FIG. Further, if the search area changes from the high content area to the non-mark area by a preset number of times, it is considered that a cross mark exists inside the search area. Further, when the search area is continuously classified into the high content area up to a preset number of times, the base point cannot be detected. In this case, it is possible to stop the base point detection process and perform a process such as notifying an operator of an error message to that effect.

  Note that when the number of base points detected in the above process is three or less, for example, as shown in FIG. 7C, the top and bottom and the left end of the image area are non-mark areas, and the right end is a mark area. The end points of the cross mark exist between the mark area and the mark area (between left and right) or between the non-mark area and the non-mark area (up and down). In such a case, the distance is converged by moving the search area between the non-mark area and the mark area (between left and right), and the end point is detected.

<Detection of mark area>
Here, in order to detect the end point described above, it is first necessary to detect a mark area. FIG. 8 is a diagram illustrating an example of a mark area detection method. In FIG. 8, as an example of the detection of the mark area, first, search is performed by setting search areas at both ends in each of the Y direction and the X direction.

  For example, the rough coordinates of the intersection of the alignment marks are known or acquired in advance during the pre-processing process (a process of including a cross mark in the shooting area by correlation calculation or manual operation). In this case, the cross mark region detection method described above is applied in the vicinity of the coordinates.

  Further, when the search area is classified as a non-mark area, or when the rough coordinates of the intersection are not known, the center of the image area where the search is completed is set as a new search area as shown in FIG. Apply the cross mark area detection method. If the search area is classified as a mark area, the search is terminated. Further, when the search area is classified into the high content area, the mark area is detected by applying the above-described cross mark area detection method to the nearby image area as in FIG. 7B described above. .

<Detection of cross mark end>
Next, an example of detecting the mark end from the detected mark area will be described with reference to the drawings. FIG. 9 is a diagram illustrating an example of detection of a cross mark end. When detecting the cross mark end, for example, as shown in FIGS. 9A and 9B, the cross mark end point is detected by converging the distance between the mark area and the non-mark area in the image area. In this case, an image region near the middle between the mark region and the non-mark region is set as a search region, and the above-described cross mark region detection method is applied.

  If the search area between the mark area and the non-mark area described above is classified as a mark area, the image area near the middle of the mark area and the previously used non-mark area is set as the search area. To explore.

  If the intermediate search area between the mark area and the non-mark area is classified as a non-mark area, an image area near the intermediate area between the non-mark area and the previously used mark area is set as the search area. To explore.

  Specifically, as shown in FIG. 9A, when there is a non-mark area and a mark area, an intermediate area is an area to be searched next, and the area is classified as a mark area. Further, the intermediate region is the region to be searched next.

  For example, as shown in FIG. 9B, when the area to be searched for includes a high content area, the search area is shifted to the non-mark area side by a predetermined width and further softened. An intermediate area is an area to be searched next after the area.

  Here, FIG. 10 is a diagram for explaining the contents of the shift processing when the search region is classified into the high content rate region. As shown in FIG. 10, when the search area is classified as a high content area, the search area is shifted in the direction away from the search area (two directions), and the above-described cross mark area detection method is applied. Further, when a mark area or a non-mark area is detected in any one of the shifted areas as shown in FIGS. 10A and 10B, the search area shift is ended, and the above-mentioned area is used as a reference. A convergence process of the distance between the mark area and the non-mark area is performed.

  Here, when the detection of the vicinity of the cross mark end point by the convergence of the distance between the mark area and the non-mark area is repeatedly performed, the search areas are converged at both ends of the cross mark. When the distance between the mark area and the non-mark area is sufficiently close, in the mark area at that time, for example, the same process as the process shown in FIG. 7A described above is performed to detect the base point. Similarly, other base points are obtained, and thereby a total of four base points can be obtained.

  In the above-described processing, two base points may be detected in a certain region. Here, FIG. 11 is a diagram illustrating an example when two base points are detected. As shown in FIG. 11, the convergence process as described above is also performed when two mark areas are detected in the search area, and two marks are detected even if the distance between the non-mark area and the mark area is sufficiently close. When a region is detected, two base points are determined therefrom.

  In addition, when the number of mark areas becomes one in the search process, the same base point is detected by performing the same convergence process between the mark area where two marks exist and the mark area where one mark exists. To do.

<Acquisition of alignment mark intersection by line (straight line)>
When at least four base points are detected by the above-described method, a straight line is obtained from the coordinates of the base point using a straight line equation using two of the detected base points, and the intersection point of the alignment mark is obtained from the intersection point of the straight lines. Note that the intersection of the alignment marks exists within a coordinate range surrounded by four base points.

  Here, FIG. 12 is a diagram illustrating an example of a set pattern of line segments that can be taken by four base points. As shown in FIG. 12, when the respective base points 21-1 to 21-4 are obtained for the alignment mark 11 displayed in the screen 10, the line segment pattern is the line segment 22 in FIG. -1, 22-2, line segments 22-3 and 22-4 in FIG. 12B, and line segments 22-5 and 22-6 in FIG. 12C.

  Here, when calculating the mark intersection point, by selecting the intersection point within the coordinate range surrounded by the four base points among the three line segment patterns of FIGS. 12A to 12C, The intersection of the marks is determined (FIG. 12 (b)).

  By detecting the alignment mark in this manner, the position detection accuracy of the alignment mark can be improved. Further, when the camera and the alignment mark have a predetermined angle, specifically, the cross point of the cross mark can be detected even when the angle is not perpendicular to the cross line of the cross mark shown in the image.

  Furthermore, even when all of the alignment marks are not displayed in the image, for example, in the case of a cross mark, it is possible to detect an accurate position because at least the position of the intersection is included in the image. .

<Laser processing apparatus having an image processing apparatus using the position detection method in the present invention>
Next, as an example of an image processing apparatus that performs the alignment mark position detection method described above, a laser processing apparatus including the image processing apparatus will be described with reference to the drawings. In the laser processing apparatus described below, an example in which the position of the alignment mark provided on the processing target is detected will be described. However, the present invention is not limited to this, and for example, to adjust the stage to an appropriate position. The present invention can also be applied to the case of detecting the position of an alignment mark provided on a mask for making the cross section of the laser beam provided on the stage a predetermined shape.

  Here, a functional configuration example of the laser processing apparatus according to the present invention will be described with reference to the drawings. FIG. 13 is a diagram showing a configuration example of a laser processing apparatus applied to the present invention. A laser processing apparatus 30 shown in FIG. 13 includes a laser oscillator 31, an optical system unit driving unit 32, an optical system unit 33, a galvano driving unit 34, a galvano mirror 35, a half mirror 36, a camera 37, and an image. The processing device 38, the stage 39, the display means 40, and the control means 41 are comprised.

The laser oscillator 31 emits a pulsed laser beam having a predetermined intensity at a predetermined timing based on a control signal obtained from the control means 41. Here, for example, a CO ₂ laser, an excimer laser, a YAG laser, or the like can be used as the laser light in the present embodiment, but the type of the laser light in the present invention is not limited to this, for example, the processing object 42 It can be arbitrarily selected depending on various processing conditions such as material, thickness, and what kind of processing (annealing, drilling, etc.) is performed.

  Based on the control signal obtained from the control means 41, the optical system unit drive means 32 condenses the laser light passing through the optical system unit 33 at a predetermined position, so that the position of the optical system unit 33 is set in the optical axis direction ( Move in the Z-axis direction). Note that, when a plurality of optical system lenses are present in the optical system unit 33, the optical system unit driving unit 32 can also move the position of at least one lens selected from these lenses.

  The optical system unit 33 is positioned at a predetermined position by the optical system unit driving means 32, and condenses the laser light passing through the optical system unit 33 at the predetermined position. Specifically, the optical system unit 33 has at least one optical system lens such as an fθ lens or a condensing lens, condenses (focuses) the laser light from the laser oscillator 31, and forms a glass substrate or a resin. A predetermined image is formed on the irradiation surface of the workpiece 42 such as a substrate.

  Based on a control signal from the control means 41, the galvano driving means 34 moves a galvano mirror 35, which is an example of a mirror having an inclination in a predetermined direction with respect to the optical axis of the laser beam from the optical system unit 33, to a predetermined position. This is mirror driving means for irradiating a predetermined processing position of the processing object 42 with laser light that has been moved and settled to the position and changed its direction by reflection by the galvanometer mirror 35. That is, the galvano driving means 34 moves the galvanometer mirror 35 to a predetermined position at a predetermined timing, thereby changing the irradiation position of the laser beam with respect to the horizontal direction (XY direction) of the processing surface of the processing object 42. Laser processing such as annealing can be performed by irradiating laser light while scanning a predetermined processing position of the object. Note that the galvano driving means 34 and the galvano mirror 35 may be provided with two mechanisms for scanning in the X direction and for scanning in the Y direction.

  The half mirror 36 is a mirror for photographing the surface of the workpiece 42 and the stage 39 from the optical axis of the laser light emitted from the laser oscillator 31 by the camera 37 as an imaging unit. That is, the Her mirror 36 transmits the laser beam from the laser oscillator 31 completely and causes the camera 37 to image the surface of the workpiece 42 and the stage 39.

  The camera 37 photographs the processing object 42 and the surface of the stage 39 from the half mirror 36, and holds the at least one processing object alignment mark 43 and / or the processing object 42 present on the processing object 42. At least one stage alignment mark 44 is obtained. The position, number, shape, and the like of the workpiece alignment mark 43 and the stage alignment mark 44 are not particularly limited. Further, the shapes of the workpiece alignment mark 43 and the stage alignment mark 44 may be the same or different.

  Note that the optical axis of the laser light emitted from the laser oscillator 31 is normally positioned at the center of the workpiece 42 and the stage 39 in order to efficiently swing the laser light by the galvano driving means 34. Therefore, by acquiring an image from the same position as the optical axis of the laser beam emitted from the laser oscillator 31, a surface image from the center of the workpiece 42 and the stage 39 can be easily acquired. The camera 37 outputs the captured image to the image processing device 38.

  The image processing device 38 acquires position (intersection) information of the workpiece alignment mark 43 and / or the stage alignment mark 44 from the image acquired from the camera 37 by the above-described alignment mark position detection method. Further, the image processing device 38 outputs the acquired positional information of the processing object alignment mark 43 and the stage alignment mark 44 to the control means 41.

  The stage 39 holds the workpiece 42 at a predetermined position by, for example, vacuum suction. Further, the display means 40 displays on the image processing statuses and processing results in laser processing, detection results relating to the above-described alignment mark position detection, and various messages generated by the control means 41 when the detection process is completed or stopped. To do.

  Furthermore, the display means 40 can display, for example, initial settings in laser processing, a processing state, processing results, and the like. Thereby, a user such as an operator can easily obtain the position of the alignment mark from the display screen. In addition, the machining state and the machining result can be easily grasped from the screen.

  The control means 41 controls drive in the laser oscillator 31, the optical system unit drive means 32, and the galvano drive means 34. Specifically, the control means 41 moves the galvano mirror 35 by the galvano driving means 34 based on the alignment mark position detection result obtained from the image processing device 38, and brings the processing object 42 into a predetermined position. Move.

  Further, the control unit 41 adjusts the optical system unit driving unit 32 in the optical axis (Z-axis) direction, and performs focusing on the workpiece 42 by the optical system unit 33. The control means 41 controls the emission timing of the laser light in the laser oscillator 31 and the drive timing of the galvano drive means 36 for moving the reflection surface of the galvano mirror 35 in a predetermined direction during processing.

  Further, the control means 41 can perform image acquisition by the camera 37, position detection of the alignment mark by the image processing device 38, and position correction according to the detection result every time the processing target is changed. It can also be performed during intervals, at regular intervals, during maintenance, or during operation preparation of the apparatus.

  Further, the control means 41 positions the optical system unit 33, the galvanometer mirror 35, and the workpiece 42 in a predetermined position corresponding to a preset processing condition, and then irradiates a laser beam with a predetermined intensity. Thereby, since the workpiece 42 can be positioned with high accuracy, high-precision laser processing can be realized.

  As described above, according to the present invention, the position detection accuracy of the alignment mark can be improved. Thereby, since the exact position of objects, such as a stage, an object to be processed, and a mask, can be grasped, highly accurate laser processing can be realized. Specifically, by detecting the base point from a part of the alignment mark included in the photographed image and searching for the alignment mark from the detected base point, the alignment accuracy is higher than the conventional pattern matching or projection method. The correct position.

  The alignment mark detection method described above is not limited to the laser processing apparatus, and can be widely applied to the field of other positioning mechanisms such as an exposure apparatus.

  The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to such specific embodiments, and various modifications can be made within the scope of the gist of the present invention described in the claims. Can be changed.

It is a figure of an example for demonstrating the projection method which is a prior art. It is a figure of an example for demonstrating the case where the projection method cannot be applied. It is a figure for demonstrating the base point of the alignment mark in this invention. It is a flowchart which shows an example of the cross point detection procedure of a cross mark. It is a figure of an example for demonstrating the area | region detection method of a cross mark. It is a figure for demonstrating a high content rate area | region. It is a figure for demonstrating the search and detection method of a base point. It is a figure of an example for demonstrating the detection method of a mark area | region. It is a figure which shows an example of the detection of a cross mark end. It is a figure for demonstrating the content of the shift process when a search area | region is classified into the high content rate area | region. It is a figure which shows an example in case two base points are detected. It is a figure which shows the example of the set pattern of the line segment which four base points can take. It is a figure which shows the example of 1 structure of the laser processing apparatus applied to this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image 11 Alignment mark 12,21 Base point 13 Straight line 22 Line segment 30 Laser processing apparatus 31 Laser oscillator 32 Optical system unit drive means 33 Optical system unit 34 Galvano drive means 35 Galvano mirror 36 Half mirror 37 Camera 38 Image processing apparatus 39 Stage 40 Display means 41 Control means 42 Work object 43 Work object alignment mark 44 Stage alignment mark

Claims (7)

In the alignment mark position detection method for detecting at least one alignment mark present on the object from an image obtained by photographing the object to be aligned,
A mark acquisition step of acquiring a preset alignment mark from the image;
A base point detection step of detecting at least four base points from the alignment mark acquired by the mark acquisition step;
A line segment acquisition step of acquiring a plurality of line segments from the base point detected by the base point detection step;
A position detection step of detecting a position of the alignment mark from an intersection of the line segments obtained by the line segment acquisition step. The base point detection step includes
It is a point that exists in a region constituting the alignment mark, the base points are separated from each other by a predetermined distance, and further, a point where three or more points are not arranged on the same straight line is detected as a base point. The position detection method according to claim 1. The base point detection step includes
The base point is detected from the contact portion when the alignment mark is a cross mark and a contact portion between the edge of the image capturing area of the image and the alignment mark exists. Position detection method. The base point detection step includes
4. If at least four base points are not detected, a part of a straight line constituting the alignment mark is searched by a preset search area, and the base point is detected from the search result. The position detection method according to any one of the above. The base point detection step includes
5. The method according to claim 1, wherein when three or more base points are detected on the same straight line, the points at both ends are left as base points and the other points are deleted. 6. Position detection method. The mark acquisition step includes:
2. The search area set for searching for the alignment mark in the image is classified into one of a mark area, a non-mark area, and a high content ratio area. 6. The position detection method according to any one of items 1 to 5. A laser processing apparatus that performs processing by irradiating a laser beam to a positioned workpiece by the alignment mark position detection method according to any one of claims 1 to 6.
An optical system unit driving means for moving the optical system unit in the optical axis direction with respect to the laser light and irradiating the laser light imaged on the workpiece;
Mirror driving means for irradiating a predetermined position of the object to be processed by scanning the laser beam from the optical system unit by moving a mirror;
Imaging means for photographing the workpiece and / or the surface of the stage holding the workpiece;
An image processing device for detecting the position of the alignment mark with respect to the image captured by the imaging means;
A laser processing apparatus comprising: a control unit for adjusting a position of a laser beam scanned by the mirror driving unit based on a position detection result of the alignment mark obtained by the image processing apparatus.

Images